Add functionality to use GNINA pre-trained models

[RMeli]

Weights of the pre-trained GNINA models were provided in #34 by @drewnutt , getting a steps closer towards #33 .

The following needs to be implemented in order to have the weights readily available:

• Automatic loading weights into the current models
  • default2017 Load pre-trained GNINA default2017 and default2018 models #36
  • default2018 Load pre-trained GNINA default2017 and default2018 models #36
  • dense Add support for dense models and GNINA default model #44
• Support ensemble of models Add ensemble of GNINA models #38
  • Support GNINA default model Add support for dense models and GNINA default model #44
• Publish package on PyPI and conda-forge
  • Change name to gninatorch Rename package as gninatorch #37

This additional functionality could be implemented in gnina.py and follow the GNINA command line as closely as possible. In addition to .types files, it would be good to use molgrid.openbabel to provide the --receptor and --ligand options.